On Tue, 19 Jan 1999, Nicolas Welte wrote: > Since I have no experience at all with techniques to reduce this truth > table to a set of equations, I want to leave this task to others. Maybe > there will be a little competition who solves this first. First of all, thanks for doing this! Soon there will be one secret less about Commodore computers. I have the experience and tools, but I lack the time. I reduced the CHAROM equations from 245765 to 84 using a BDD-based tool (Binary Decision Diagrams), but the equations could probably be reduced further. BDDs are sensitive to the order in which the variables are introduced in the formulas. > The file is only available from my own server right now, but I'll upload > it to funet as well. It's now available from /pub/cbm/firmware/computers/c128/8721.zip. > BTW, I can still be talked into providing a more complete dump, with 26 > inputs and 16 outputs. This is not such a big task. Now you have wired the 2 untied inputs to fixed values, right? Then produce 3 other dumps using the 3 other values (move the two inputs from their current setting to the other setting, one at a time (e.g. 00->01->11->10), and you don't need to solder so much). The dumps can then be catenated together. I have access to a 512 MB computer, so there shouldn't be no swapping problems when I generate the equations from the dumps. With rather simple rewiring, you could produce 4 dumps of the remaining 4 output bits as well. My simple program is designed for byte-aligned truth tables, so don't try to pack 2 samples to one byte. > And I'm not going to dump all 18 output bits, because the use of the > extra two is pretty obvious. It may be, but I'd like to see them anyway. Marko - This message was sent through the cbm-hackers mailing list. To unsubscribe: echo unsubscribe | mail cbm-hackers-request@dot.tcm.hut.fi.
Archive generated by hypermail 2.1.1.